448 PROCEEDINGS OF THE AMERICAN ACADEMY. 



holism. Furthermore, while the supply of food in cultures and the 

 constitution of the medium, especially with reference to its salt content, 

 are normal, no maxima and mimima of division occur. This condition 

 never lasts indefinitely, and both observation and chemical examination 

 show that there is a definite course of changes natural to cultures from 

 the period of their beginning with a food supply to that of its exhaustion. 

 In an undisturbed culture the decline of Stentor as well as its initial 

 rapid multiplication is due to metabolic couditions. In the present ex- 

 periments old cultures were frequently revived, and rapidly multiplying 

 ones were sometimes brought to suilden decline, by an apparently slight 

 change in the nutritive conditions of the medium. The experiments on 

 salt media described in this work show the pronounced sensitiveness of 

 these animals to this factor. The response of Stentor to these conditions 

 points to metabolic exigency as the efficient factor. To demonstrate ex- 

 perimentally, i. e. by more eihcient means than observation of cultures, 

 that periodicity of division occurs in a protozoan, two conditions at 

 least must be fulfilled. First, the experimenter must know the exact 

 composition with reference to salts, food, etc., of the media to which he 

 subjects the animals. Such promiscuous mixtures as hay infusion of 

 wiknoio?i composition will not suffice. Since frequent chemical analyses 

 are impracticable, it will be necessary to construct by trial artificial 

 media of known composition. The second requirement is that the ex- 

 perimenter should determine the normal adjustments of the animal to the 

 different factors of the proposed media. Without this knowledge it is 

 impossible to ascertain whether observed increase and decrease of multi- 

 plication is due to simple metabolic exigency or to an inherent tendency 

 to cyclic division. A begiiniing towards meeting both requirements 

 above indicated has been made for Stentor by experiments subsequently 

 described. 



In the preparation of reagents such precautions were taken as were 

 appropriate to the end in view, and hence varied somewhat in different 

 experiments. In general this part of the work was made to rest as much 

 as possible upon objective demonstration. Reagents were either tested, 

 even though their labels might be those of reliable manufacturers, or 

 they were prepared personally. The measures used for volumetric pur- 

 poses were calibrated by means of weighed quantities of liquid, using 

 Sutton's ('96, p. 25) table for air weighing with brass weights. 



Standard solutions were carefully made, and consistency among dilute 

 solutions, chlorides, etc., maintained. Dilutions with water from an 

 originally strong solution were made by weight with reference to the 



